{"id":"https://openalex.org/W7161052565","doi":"https://doi.org/10.48550/arxiv.2605.12031","title":"Resilient Vision-Tabular Multimodal Learning under Modality Missingness","display_name":"Resilient Vision-Tabular Multimodal Learning under Modality Missingness","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161052565","doi":"https://doi.org/10.48550/arxiv.2605.12031"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12031","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12031","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.12031","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5065872543","display_name":"Camillo Maria Caruso","orcid":"https://orcid.org/0000-0002-7320-4173"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Caruso, Camillo Maria","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136037753","display_name":"Valerio Guarrasi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guarrasi, Valerio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136054165","display_name":"Paolo Soda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Soda, Paolo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5927000045776367,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.5927000045776367,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.09009999781847,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.05869999900460243,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/missing-data","display_name":"Missing data","score":0.6624000072479248},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.6047000288963318},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.5939000248908997},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5099999904632568},{"id":"https://openalex.org/keywords/multimodal-learning","display_name":"Multimodal learning","score":0.5042999982833862},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.49000000953674316},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.4697999954223633},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.45019999146461487},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.42669999599456787},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.42309999465942383}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7609000205993652},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6894000172615051},{"id":"https://openalex.org/C9357733","wikidata":"https://www.wikidata.org/wiki/Q6878417","display_name":"Missing data","level":2,"score":0.6624000072479248},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.6047000288963318},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.5939000248908997},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5278000235557556},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5099999904632568},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.5042999982833862},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.49000000953674316},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4697999954223633},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.45019999146461487},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.42669999599456787},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42309999465942383},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.41990000009536743},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.40869998931884766},{"id":"https://openalex.org/C58041806","wikidata":"https://www.wikidata.org/wiki/Q1660484","display_name":"Imputation (statistics)","level":3,"score":0.3749000132083893},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.3546000123023987},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3377000093460083},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.33399999141693115},{"id":"https://openalex.org/C2780695315","wikidata":"https://www.wikidata.org/wiki/Q3799040","display_name":"Unobservable","level":2,"score":0.3140000104904175},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3125},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C4441509","wikidata":"https://www.wikidata.org/wiki/Q6418787","display_name":"Multimodal therapy","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C184297639","wikidata":"https://www.wikidata.org/wiki/Q177765","display_name":"Biometrics","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.26089999079704285},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C2778459887","wikidata":"https://www.wikidata.org/wiki/Q6787865","display_name":"Matrix completion","level":3,"score":0.2556999921798706},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12031","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12031","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.12031","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12031","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"deep":[1],"learning":[2,62],"has":[3],"shown":[4],"strong":[5],"potential":[6],"in":[7,37,185],"medical":[8,17],"applications":[9],"by":[10],"integrating":[11],"heterogeneous":[12],"data":[13,147,161],"sources":[14],"such":[15],"as":[16],"images":[18],"and":[19,44,88,101,113,118,182,211,221],"structured":[20,159],"clinical":[21,39,160],"variables.":[22],"However,":[23],"most":[24],"existing":[25],"approaches":[26],"implicitly":[27],"assume":[28],"complete":[29],"modality":[30,69,99,187],"availability,":[31],"an":[32],"assumption":[33],"that":[34,131,218],"rarely":[35],"holds":[36],"real-world":[38],"settings":[40],"where":[41],"entire":[42],"modalities":[43,114,135],"individual":[45],"features":[46],"are":[47,95,227],"frequently":[48],"missing.":[49],"In":[50],"this":[51],"work,":[52],"we":[53,125],"propose":[54],"a":[55,84,86,89,127],"multimodal":[56,90,191,231],"transformer":[57],"framework":[58],"for":[59,164],"joint":[60,225],"vision-tabular":[61],"explicitly":[63],"designed":[64],"to":[65,141,192,229],"operate":[66],"under":[67,145],"pervasive":[68],"missingness,":[70],"without":[71],"relying":[72],"on":[73,153],"imputation":[74],"or":[75],"heuristic":[76],"model":[77,140],"switching.":[78],"The":[79],"architecture":[80],"integrates":[81],"three":[82],"components:":[83],"vision,":[85],"tabular,":[87],"fusion":[91,105,223],"encoder.":[92],"Unimodal":[93],"representations":[94],"weighted":[96],"through":[97],"learnable":[98],"tokens":[100,112],"fused":[102],"via":[103],"intermediate":[104,222],"with":[106,158,171,224],"masked":[107],"self-attention,":[108],"which":[109],"excludes":[110],"missing":[111],"from":[115,162],"information":[116,144],"aggregation":[117],"gradient":[119],"propagation.":[120],"To":[121],"further":[122,216],"enhance":[123],"resilience,":[124],"introduce":[126],"modality-dropout":[128],"regularization":[129],"strategy":[130],"stochastically":[132],"removes":[133],"available":[134],"during":[136],"training,":[137],"encouraging":[138],"the":[139,154,200],"exploit":[142],"complementary":[143],"partial":[146],"availability.":[148],"We":[149],"evaluate":[150],"our":[151],"approach":[152],"MIMIC-CXR":[155],"dataset":[156],"paired":[157],"MIMIC-IV":[163],"multilabel":[165],"classification":[166],"of":[167],"14":[168],"diagnostic":[169],"findings":[170],"incomplete":[172],"annotations.":[173],"Two":[174],"parallel":[175],"systematic":[176],"stress-test":[177],"protocols":[178],"progressively":[179],"increase":[180],"training":[181],"inference":[183],"missingness":[184,198],"each":[186],"separately,":[188],"spanning":[189],"fully":[190,193],"unimodal":[194],"scenarios.":[195],"Across":[196],"all":[197],"regimes,":[199],"proposed":[201],"method":[202],"consistently":[203],"outperforms":[204],"representative":[205],"baselines,":[206],"showing":[207],"smoother":[208],"performance":[209],"degradation":[210],"improved":[212],"robustness.":[213],"Ablation":[214],"studies":[215],"demonstrate":[217],"attention-level":[219],"masking":[220],"fine-tuning":[226],"key":[228],"resilient":[230],"inference.":[232]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-14T00:00:00"}